demo.git - Vide-coded 64k demo system

Age	Commit message (Expand)	Author
19 hours	CNN v2 training: Refactor train_cnn_v2_full.sh for maintainability	skal
21 hours	CNN v2 training: Add --grayscale-loss option for luminance-based loss computa...	skal
22 hours	CNN v2 training: Expose all parameters as CLI options	skal
22 hours	CNN v2 full pipeline: Add --mip-level option	skal
25 hours	Add --export-only option to train_cnn_v2_full.sh	skal
26 hours	Fix train_cnn_v2_full.sh for updated API	skal
26 hours	CNN v2: Refactor to uniform 12D→4D architecture	skal
30 hours	Add weights/ subdirectory to workspaces for CNN training outputs	skal
31 hours	Refactor: Reorganize workspaces and remove assets/ directory	skal
31 hours	Refactor: Move application entry points to src/app/	skal
2 days	Update docs and help messages for CNN v2 completion	skal
2 days	Add --validate mode to training script	skal
2 days	Refine training script output and validation	skal
2 days	CNN v2: storage buffer architecture foundation	skal
2 days	CNN v2: Patch-based training as default (like CNN v1)	skal
2 days	Fix: CNN v2 training - handle variable image sizes	skal
2 days	CNN v2: parametric static features - Phases 1-4	skal
2 days	CNN v2: parametric static features - design doc	skal
5 days	refactor: Reorganize tests into subsystem subdirectories	skal
5 days	feat: Add headless mode for testing without GPU	skal
5 days	feat: Implement Task #76 external library size measurement	skal
5 days	refactor: Move .spec audio assets to assets/final/music subdirectory	skal
7 days	test: Add HTML Gantt chart output test for seq_compiler	skal
7 days	test: Add Gantt chart output test for seq_compiler	skal
7 days	feat(build): Add FINAL_STRIP mode for maximum size optimization	skal
7 days	fix(coverage): Explicitly disable STRIP_ALL during coverage runs	skal
7 days	fix(coverage): Handle moved files and clean stale coverage data	skal
7 days	fix(ci): Update verification script to catch tool compilation failures	skal
8 days	feat(audio): Add RMS normalization to spectool for consistent sample loudness	skal
10 days	asset location cleanup	skal
10 days	feat(tooling): Add directory filtering to coverage report script (Task #46)	skal
10 days	feat(tooling): Implement code coverage reporting (Task #44)	skal
11 days	refactor: Task #20 - Platform & Code Hygiene	skal
11 days	build: Enable parallel compilation in build scripts	skal
11 days	feat(assets): Add new drum samples and improve conversion script	skal
12 days	fix(shader): Correct WGSL loop syntax in calc_shadow	skal
12 days	add new tasks	skal
12 days	feat(build): Add check_all script and optimize spectool	skal
14 days	refactor: move generated asset files to src/generated/	skal
14 days	fix: Cross-compilation and style compliance	skal
2026-01-31	update session with mix fixes	skal
2026-01-31	update state in .md files	skal
2026-01-31	Finalize Windows port with size analysis reporting and updated docs	skal
2026-01-31	Add Windows cross-compilation support (MinGW) and emulation (Wine)	skal
2026-01-31	Chore: Remove trailing whitespaces across the codebase	skal
2026-01-28	chore(scripts): Update gen_assets.sh to include test asset generation	skal
2026-01-28	fix(assets): Update gen_assets.sh paths	skal
2026-01-28	fix(crunch): Use strip/gzexe on macOS instead of UPX	skal
2026-01-28	feat(crunch): Add UPX-based binary packer script	skal
2026-01-28	refactor(gpu): Integrate WebGPU via system wgpu-native and glfw3webgpu	skal